Statistical Inference

INTRODUCTION TO STATISTICAL INFERENCE The intent of the mineral appraiser in collecting samples from a mineral deposit is to arrive at accurate conclusions concerning the size and value of the deposit. In other words, he wants to derive reliable inferences about the total mineral population from the collected sample data, which (he hopes) is representative of the whole. The discussion of sampling theory in Chapter 2 was oriented toward assuring that our sampling techniques and sampling units would indeed be representative of the total mineral population. Chapter 3 described the most common statistical parameters for defining distribution in various populations. Chapter 4 gave some insight into theoretical distributions of random variables and the probabilities associated with them. The intent of this chapter is to derive inferences from the preceding information about the population mean and the population variance. Knowing something about the normal distribution, and assuming our samples represent a roughly normal population, we can make inferences or logical deductions about the estimated interval or the limits of values which can be assigned to a mineral population. ESTIMATION OF POPULATION PARAMETERS In the application of statistics to mineral deposits it is never possible to know the actual values of the parameters or the theoretical distribution of values within the deposit for which our data are samples. Even after the deposit has been mined out it is questionable if the true value of the parameters can be retrieved accurately, due to dilution, mining, and processing losses. Therefore, our problem is to determine what function of our sample values we should use to estimate the unknown parameters of greatest interest to us-namely, the population mean and the population variance of the deposit. Since we cannot know